Primary Squamous Cell Carcinoma Arising From an Epithelium ... · Squamous cell carcinoma is a rare neoplasm of the lacrimal gland and may be either primary or secondary in origin

Brief Reports

Primary Squamous Cell CarcinomaArising From an Epithelium-LinedCyst of the Lacrimal GlandGrant W. Su, M.D.*, Michael Patipa, M.D.† andRamon L. Font, M.D.*

Abstract: A 66-year-old man presented with a slowlyenlarging, nontender left orbital mass of 2 months’duration. CT and MRI showed a left lacrimal glandmass with enhancement and internal irregularity ofcystic structures. Histopathologic analysis of the bi-opsy specimen revealed a squamous cell carcinomaarising from an epithelium-lined cyst. The patientunderwent left orbital exenteration followed by radi-ation treatment. No evidence of tumor recurrencewas observed after a follow-up of 30 months. Webelieve this primary squamous cell carcinoma mayhave arisen either from preexisting lacrimal duct cyst(dacryops) with areas of squamous metaplasia or, lesslikely, from a choristomatous epithelium-lined cyst ofthe lacrimal gland. Although rare, this entity shouldbe included in the differential diagnosis of cysticlesions of the lacrimal gland.

S quamous cell carcinoma (SCC) is a rare neoplasmof the lacrimal gland, accounting for less than 2%

of primary malignant epithelial tumors.1 In most reportedcases, it arises from the malignant transformation of apleomorphic adenoma (benign mixed tumor).2–7 Well-documented cases demonstrating features of pure SCCarising from the lacrimal gland are rare.8 We describe theclinical, radiologic, and histopathologic findings of aprimary SCC arising from an epithelium-lined cystwithin the lacrimal gland.

CASE REPORT

A 66-year-old man complained of double vision of 9months’ duration after cataract surgery of the left eye. Healso noted a slowly enlarging mass of 2 months’ durationin the upper lateral aspect of his left orbit. The patientreported a sense of orbital pressure but had no pain.Ocular history was significant for uncomplicated cataractsurgery in both eyes. There was no history of penetratingocular trauma, infection, or radiation treatment. Themedical history was significant for excision of severalcutaneous basal cell and squamous cell carcinomas of hisextremities. There was no clinical evidence of basal cellnevus/Gorlin syndrome or Bazex syndrome and no fam-ily history of cancer.

Best-corrected visual acuity was 20/20 OU. There wasa 1-cm, nontender, soft tissue mass in the left uppereyelid and lateral orbit (Fig. 1). Exophthalmometry read-ings were 14 mm OD and 17 mm OS. Extraocularmovements showed marked limitation of elevation andabduction OS. No palpable lymphadenopathy waspresent. The remainder of his ocular examination, in-cluding slit-lamp and dilated ophthalmoscopic examina-tion, was normal.

CT revealed a left lateral orbital mass without involve-ment of bone. MRI demonstrated a mass lesion centeredin the region of the left lacrimal gland measuring 2.5 �2.5 � 1.5 cm in greatest dimensions (Fig. 2). On T1-weighted imaging, the lesion showed low signal intensityand was isointense to muscle or gland parenchyma. OnT2-weighted imaging, the lesion was heterogenous, withareas of hyperintense signal. Gadolinium contrast imag-ing showed enhancement of the lesion with tiny areas ofnonenhancement, consistent with either cystic changes ornecrosis. The margins of the lesion were ill-defined, withlocal infiltration of the extraconal fat. The mass moldeditself along the temporal margin of the globe, which wasslightly displaced medially. The lesion merged with theinsertion of the lateral rectus muscle. No extension in thetemporalis muscle was noted. The bone structures ap-peared intact. MRI of the brain was normal. Metastaticworkup, including bone scans and CT of the head, neck,chest, abdomen, and pelvis, were unremarkable.

The patient underwent excisional biopsy of the leftlacrimal gland, and the frozen sections revealed an epi-thelial lacrimal gland tumor. The lacrimal gland wasexcised, but residual tumor was present at the posteriororbital margin and was not resectable due to its posteriorlocation. The initial anatomic pathology was an infiltrat-ing moderately well-differentiated SCC of the lacrimalgland. Due to the diagnosis and residual tumor, which

*Department of Ophthalmology, Cullen Eye Institute, and the De-partment of Pathology, The Methodist Hospital, Baylor College ofMedicine, Houston, Texas, U.S.A.; and †Oculoplastic and OrbitalConsultants, P.A., West Palm Beach, Florida, U.S.A.

Accepted for publication March 21, 2005.Supported in part by grants from the Retina Research Foundation,

Houston, Texas, and Research to Prevent Blindness, Inc., New York,New York. Dr. Font is the recipient of the Senior Investigator Awardfrom Research to Prevent Blindness, Inc., New York, New York.

Presented at the 35th Annual American Society of OphthalmicPlastic and Reconstructive Surgery 2004 Symposium, New Orleans,Louisiana, October 22 to 23, 2004.

Address correspondence and reprint requests to Dr. Ramon L. Font,Ophthalmic Pathology Laboratory, Cullen Eye Institute, BaylorCollege of Medicine, Houston, TX 77030, U.S.A. E-mail: [email protected]

DOI: 10.1097/01.iop.0000176263.07921.22

Ophthalmic Plastic and Reconstructive SurgeryVol. 21, No. 5, pp 383–404©2005 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.

383

was not resectable with clear margins, a subsequentorbital exenteration was performed. The bone from thesuperior orbital rim lateral to the superior orbital neuro-vascular complex and extending to the lateral canthuswas resected. The entire orbit was exenterated en bloc.The final pathology margins were free of tumor. Thepatient subsequently underwent radiation treatment andwas reported to be alive and well without evidence ofrecurrence 30 months after initial surgery.

Pathologic Findings

Histopathologic examination revealed a chronic scleros-ing dacryoadenitis and areas of lymphoid hyperplasia asso-ciated with infiltrating islands of moderately well-differen-tiated SCC arising from an epithelium-lined, multiloculatedcyst (Fig. 3). Immunohistochemical markers for pankeratindemonstrated that the tumor cells were moderately immu-noreactive. Islands of invasive squamous cell carcinomawere located adjacent to the inflamed lacrimal gland (Fig.4). In some areas, the tumor was well keratinized andformed masses of keratin pearls. No evidence of a preex-isting benign mixed tumor of the lacrimal gland waspresent. On the basis of the exenteration specimen, allsurgical margins were tumor-free.

DISCUSSION

Squamous cell carcinoma is a rare neoplasm of thelacrimal gland and may be either primary or secondary inorigin. Primary SCC may arise from squamous transfor-mation in a pleomorphic adenoma (benign mixed tumor).In addition, primary SCC may develop from the malig-nant degeneration of epithelium-lined cysts, includinglacrimal duct cysts (dacryops) and choristomatous cysts.SCC may also occur secondarily from postsurgical orposttraumatic implantation cysts and metastatic spread tothe lacrimal gland.

In our case, we believe the SCC probably resultedfrom the malignant transformation of either preexistinglacrimal duct cyst (dacryops) or, less likely, from achoristomatous epithelium-lined cyst. The cyst may havedeveloped in association with chronic nongranulomatousdacryoadenitis, where the inflammatory process led toductal ectasia and squamous metaplasia of the ductalelements. Invasive SCC developed from the malignantdegeneration of the squamous-lined cyst. Alternatively,the SCC may have resulted from the malignant transfor-mation of a choristomatous epithelium-lined cyst. Al-

FIG. 2. Coronal MRI shows a mass with several cystic areasinvolving the lacrimal gland.

FIG. 1. Mass involving the left lacrimal gland region causes anS-shaped deformity of the left upper eyelid.

FIG. 3. Left, Epithelial cyst lined by stratified squamous epithelium (pankeratin, magnification 26). Right, Another epithelium-lined cystwith infiltrating cords of squamous cell carcinoma originating from cyst walls (pankeratin, magnification 16).

384 BRIEF REPORTS

Ophthal Plast Reconstr Surg, Vol. 21, No. 5, 2005

though uncommon, several well-documented cases ofinvasive SCC arising from choristomatous cysts of theorbit have been reported.9

Given the history of cutaneous squamous cell skincarcinoma, direct or metastatic extension may have oc-curred. However, there were no histopathologic or radio-graphic data to support this process. An extensive met-astatic workup was also negative. The patient had nohistory of penetrating trauma or orbital surgery to sug-gest the possibility of epithelial implantation. Further-more, no evidence of a pleomorphic adenoma (benignmixed tumor) of the lacrimal gland was found on his-topathologic examination.

Primary epithelial tumors of the lacrimal gland exhib-iting only malignant squamous elements are exceedinglyrare, with only one other well-documented case reported.Fenton et al.8 described an 80-year-old woman whopresented with an 8-month history of painless progres-sive proptosis of the left eye. CT revealed a left-sidedlacrimal gland mass without bony erosion. Surgical ex-cision of the tumor revealed a poorly differentiatedinvasive SCC of the lacrimal gland, arising from a cystlined by dysplastic squamous epithelium. No evidence ofa pleomorphic adenoma (benign mixed tumor) or meta-static process was present. Given the incomplete exci-sion, the patient subsequently received postoperativeradiation treatment. In this case,8 follow-up of 20 monthsshowed no evidence of recurrence.

Although rare, malignant degeneration of epithelialcysts should be considered in the differential diagnosis ofcystic lesions of the lacrimal gland. Primary SCC arisingfrom the malignant degeneration of epithelial cysts maynot exhibit the characteristic signs of malignant epithelialtumors of the lacrimal gland, such as pain and bonyerosion. Surgical excision followed by radiation therapymay play an important role in the treatment of primarySCC of the lacrimal gland.

REFERENCES1. Shields JA, Shields CL, Epstein JA, et al. Review: primary epithelial

malignancies of the lacrimal gland: the 2003 Ramon L. FontLecture. Ophthal Plast Reconstr Surg 2004;20:10–21.

2. Font RL, Smith SL, Bryan RG. Malignant epithelial tumors of thelacrimal gland: a clinicopathologic study of 21 cases. Arch Oph-thalmol 1998;116:613–6.

3. Wright JE, Rose GE, Garner A. Primary malignant neoplasms of thelacrimal gland. Br J Ophthalmol 1992;76:401–7.

4. Grossniklaus HE, Levine MR, Kincaid MC, et al. Squamous cellcarcinoma of the lacrimal gland. Orbit 1988;7:79–86.

5. Ni C, Cheng SC, Dryja TP, et al. Lacrimal gland tumors: aclinicopathological analysis of 160 cases. Int Ophthalmol Clin1982;22:99–120.

6. Henderson JW, Farrow GM. Primary malignant mixed tumors of thelacrimal gland. Ophthalmology 1980;87:466–73.

7. Font RL, Gamel JW. Epithelial tumors of the lacrimal gland: ananalysis of 265 cases. In: Jakobiec FA, ed. Ocular and AdnexalTumors. Birmingham, Ala: Ausculapius; 1978;787–805.

8. Fenton S, Srinivasan S, Harnett A, et al. Primary squamous cellcarcinoma of the lacrimal gland. Eye 2003;17:424–5.

9. Holds JB, Anderson RL, Mamalis N, et al. Invasive squamous cellcarcinoma arising from asymptomatic choristomatous cysts of theorbit: two cases and a review of the literature. Ophthalmology1993;100:1244–52.

Orbital Peripheral T-CellLymphoma in a ChildAdam S. Hassan, M.D.*, andVictor M. Elner, M.D., Ph.D.*†

Abstract: A 6-year-old boy with known peripheralT-cell lymphoma had progressive left periorbitalswelling and CT findings consistent with abscess ornecrosis. Biopsy revealed peripheral T-cell lymphomaand associated necrosis involving the lacrimal glandand surrounding orbital tissue. Immunohistochemi-cal and T-cell receptor gene rearrangement studiesconfirmed the diagnosis. The patient responded tolocal radiation and systemic chemotherapy. This isthe first pathologically confirmed case of orbital pe-ripheral T-cell lymphoma with subpanniculitic fea-tures.

P eripheral T-cell lymphomas (PTCL) are character-ized by infiltrates of malignant cells whose immu-

nophenotypes mimic mature T cells.1 These lymphomascomprise a wide variety of entities distinguished fromone another by their clinical and pathologic features.Involvement of extranodal sites, including skin, is acommon feature of PTCL.1 Ocular manifestations ofPTCL are rare and occur almost exclusively in adults, inwhom the eyelid is the most commonly involved.2,3

Departments of *Ophthalmology and †Pathology, University ofMichigan, Ann Arbor, Michigan, U.S.A.

Accepted for publication March 18, 2005.Address correspondence and reprint requests to Dr. Victor M. Elner,

University of Michigan, Kellogg Eye Center, 1000 Wall Street, AnnArbor, MI 48105, U.S.A. E-mail: [email protected]

DOI: 10.1097/01.iop.0000176264.30697.02

FIG. 4. Chronic dacryoadenitis with remnants of an inflamedlacrimal gland (on left) merging with infiltrating epithelial lobulesof a moderately well-differentiated squamous cell carcinoma(hematoxylin and eosin stain, magnification 40�).

385BRIEF REPORTS


Subcutaneous panniculitis-like T-cell lymphoma(SPTCL) is a type of primary extranodal PTCL that hasnot been histopathologically documented in the orbit. Wereport the only documented case of orbital involvementby SPTCL.

CASE REPORT

In August 2002, a 6-year-old boy with known PTCLinvolving the skin and subcutaneous tissues was referredfor evaluation of left periorbital swelling characterizedby diffuse redness and pain (Fig. 1). His immunocom-promise by weekly vinblastine treatment raised the pos-sibility of orbital cellulitis, even though his involvementresembled his other PTCL lesions. CT revealed preseptaland superolateral orbital infiltrates that involved thelacrimal gland and molded to the eye (Fig. 1). The lucentareas within the infiltrate were consistent with necrosisor abscess. An orbital biopsy was performed.

Histopathology revealed diffuse infiltration of lacri-

mal gland tissue by neoplastic, large lymphocytes and apolymorphous infiltrate of T -cells, B cells, histiocytes,and plasma cells (Fig. 2A). The periglandular orbitalconnective tissue revealed extensive necrosis and hem-orrhage with an infiltrate consisting almost exclusivelyof neoplastic lymphocytes, without abscess formation(Fig. 2B). Immunohistochemistry of lacrimal gland tis-sue demonstrated numerous CD3- and CD43-positiveneoplastic T cells and modest numbers of B cells stainingwith CD20 and CD79a antibodies. The periglandular,partially necrotic orbital tissue also contained numerousneoplastic CD3- and CD43-positive T cells but noCD20- or CD79a-positive B cells. Polymerase chainreaction of the specimen revealed T-cell receptor-�(TCR-�) gene rearrangement. There was no immunore-activity for Epstein-Barr small nuclear RNA (EBER),CD30/Ki-1, or CD56/CD57 nuclear killer cell markers.

To confirm that the orbital lymphoma was a manifes-

FIG. 1. A, Six-year-old boy with acute left periorbital swelling;B, coronal CT image with superolateral infiltrate containing cen-tral lucency; C, reduced left periorbital swelling after radiother-apy and chemotherapy.

FIG. 2. A, Lacrimal gland infiltrated by malignant T-lympho-cytes; B, orbital tissue with necrosis and malignant lymphocyticinfiltrate (hematoxylin and eosin stain; magnification 200�); C,skin biopsy demonstrating diffuse dermal and subdermal fatSPTCL infiltrates (hematoxylin and eosin stain; magnification25�); D, subdermal fat and periadnexal infiltrates of SPTCL inskin biopsy (hematoxylin and eosin stain; magnification 100�).

386 BRIEF REPORTS


tation of PTCL, previous skin biopsies were reviewed.At age 23 months, as the first manifestation of hisdisease, persistent, painful, red nodules appeared overthe course of 2 months. Biopsy specimens of theselesions were used to diagnose PTCL. Histopathologyrevealed large neoplastic cells with convoluted nucleiand intermingled, numerous benign histiocytes. The neo-plastic cells spared the epidermis and were present inlarge aggregates with massive extension in subdermal fat(Fig. 2C). The neoplastic cells congregated around skinappendages and blood vessels (Fig. 2D). Immunohisto-chemistry confirmed CD3 and CD43 tumor cell positiv-ity and CD68 positivity of intermingled histiocytes. RareCD20 or CD79a stained B cells were present. Concurrentchest wall masses were present at the time of initialdiagnosis, but no lymph node or bone marrow involve-ment was ever documented. Vinblastine induced tumorregression, and an autologous peripheral stem cell trans-plant was performed but failed due to persistent subclin-ical disease. Subsequent clinically evident tumor recur-rence required weekly maintenance vinblastine therapyto control PTCL.

After the diagnosis of orbital SPTCL, radiation ther-apy (36 Gy) was used to eradicate the lymphoma. Sub-sequent systemic pentostatin (10 cycles of 5 mg/m2 givendaily, for 3 days, every 3 weeks) treatment, initiated toaddress his residual systemic disease, resolved his orbitaland skin manifestations (Fig. 1). His disease has re-mained under control for 2.5 years after this regimen. Hisleft eye exhibits mild enophthalmos and a deep superiorsulcus.

DISCUSSION

Peripheral T-cell lymphomas account for up to 20% ofnon-Hodgkin lymphomas in children.4 This heteroge-nous group of T-cell lymphomas includes those withchiefly extranodal skin manifestations.4 Nevertheless, adistinctive subtype of PTCL, SPTCL,1 is rare in childrenand has only been reported in the orbit once.5 In thatcase, the presumed diagnosis of orbital SPTCL wasbased on skin biopsy and CT detection of a mass in thelacrimal gland region. To our knowledge, this is the firstpathologically confirmed case of orbital SPTCL.

Our pathologic findings of the orbital specimen werecharacteristic of SPTCL (Fig. 2), including the presenceof necrosis, infiltration by large, malignant T cells, andTCR-� gene rearrangement.6 Inasmuch as the clinicalpresentation of the lesion resembled the patient’s otherSPTCL lesions and the orbital biopsy was beneath thedermis, his periocular involvement was subpanniculitic,meeting the criteria for SPTCL. The skin biopsy speci-mens also demonstrated SPTCL characteristics reportedin children, namely, periappendiceal and perivasculartumor cell aggregates, diffuse involvement of subdermalfat, and permeation of the tumor cell aggregates by

numerous CD68� histiocytes.6 As in other cases ofSPTCL, our case was not immunoreactive for Epstein-Barr small nuclear RNA (EBER), CD30/Ki-1, or CD56/CD57 nuclear killer cell markers.6

The clinical and radiologic differential diagnosis inour case included orbital cellulitis/abscess, whose CTfindings may mimic the necrosis due to SPTCL. Biopsyof the involved tissue, however, showed only malignantT cells in areas of necrosis, inconsistent with an infec-tious pathology. It is likely that the necrosis associatedwith SPTCL was at least partially responsible for theenophthalmos and deep superior sulcus.

The clinical course of SPTCL is varied, ranging fromrapidly progressive and fatal to protracted and recurrent.5

Our patient has survived 6 years since presentation witha chronic, recurring course controlled by systemic che-motherapy and local radiation. However, he failed bonemarrow transplantation, which has been advocated.5

REFERENCES1. Fink-Puches R, Zenahlik P, Back B, et al Primary cutaneous

lymphomas: applicability of current classification schemes (Euro-pean Organization for Research and Treatment of Cancer, WorldHealth Organization) based on clinicopathologic features observedin a large group of patients. Blood 2002;99:800–5.

2. Cook BE Jr, Bartley GB, Pittelkow MR. Ophthalmic abnormalitiesin patients with cutaneous T-cell lymphoma. Trans Am OphthalmolSoc 1998;96:309–27.

3. Coupland SE, Foss HD, Assaf C, et al T-cell and T/natural killer-cell lymphomas involving ocular and ocular adnexal tissues: aclinicopathologic, immunohistochemical, and molecular study ofseven cases. Ophthalmology 1999;106:2109–20.

4. Agnarsson BA, Kadin ME. Peripheral T-cell lymphomas in chil-dren. Semin Diagn Pathol 1995;12:314–24.

5. Hung IJ, Kuo TT, Sun CF. Subcutaneous panniculitic T-cell lym-phoma developing in a child with idiopathic myelofibrosis. J Pedi-atr Hematol Oncol 1999;21:38–41.

6. Salhany KE, Macon WR, Choi JK, et al Subcutaneous panniculitis-like T-cell lymphoma: clinicopathologic, immunophenotypic, andgenotypic analysis of alpha/beta and gamma/delta subtypes.Am J Surg Pathol 1998;22:881–93.

Superior Ophthalmic VeinThrombosis in a Patient WithDacryocystitis-Induced OrbitalCellulitisNicholas J. Schmitt, M.D., Randall L. Beatty, M.D., andJohn S. Kennerdell, M.D.

Abstract: A 71-year-old-man presented with chronicleft-sided epiphora and a 5-day history of progressiveleft orbital swelling that had started with a “bump”

Allegheny General Hospital, Pittsburgh, Pennsylvania, U.S.A.Accepted for publication March 2, 2005.Address correspondence and reprint requests to Dr. Nicholas J.

Schmitt, Allegheny General Hospital, 420 East North Avenue, Suite116, Pittsburgh, PA 15212, U.S.A. E-mail: [email protected]

DOI: 10.1097/01.iop.0000176269.84949.96

387BRIEF REPORTS


on the left side of his nose. Orbital CT revealedleft-sided preseptal and postseptal inflammation,along with marked thickening of the left superiorophthalmic vein. Orbital MRI with gadolinium en-hancement and fat suppression revealed a low-inten-sity signal in the left superior ophthalmic vein, con-sistent with a superior ophthalmic vein thrombosis.There was no cavernous sinus involvement. A diag-nosis was made of left-sided dacryocystitis-inducedorbital cellulitis and superior ophthalmic vein throm-bosis. Treatment consisted of intravenous vancomy-cin, followed by early dacryocystorhinostomy andpostoperative intravenous dexamethasone. Anticoag-ulation was not used. Within 1 week after surgery, theorbital congestion had dramatically improved.Though rare, isolated superior ophthalmic veinthrombosis can be a harbinger of cavernous sinusthrombosis; therefore, early detection is the key toavoiding cavernous sinus thrombosis.

I solated septic superior ophthalmic vein thrombosis(SOVT) is uncommon in the literature.1–4 To our

knowledge, SOVT in the setting of dacryocystitis-in-duced orbital cellulitis has not been reported. In anestimated 33% to 75% of cases, septic SOVT may lead tocavernous sinus thrombosis,2 a potentially lethal compli-cation with a mortality rate of approximately 20%.1

Thus, early detection of SOVT is the key to avoidingcavernous sinus thrombosis.

Clinical differentiation of SOVT from orbital cellulitisis difficult and usually is made only after imaging theorbit. CT typically reveals an enlarged superior ophthal-mic vein as a dilated tubular S-shaped image just inferiorto the superior rectus and levator complex.3 MRI iscurrently the radiologic test of choice for confirmingSOVT because it is able to detect all stages of thrombusformation.2 Magnetic resonance venography and orbitalcolor Doppler imaging with ultrasound offer noninvasivevascular flow information, which may also confirmSOVT.2,4

CASE REPORT

An otherwise healthy 71-year-old white man pre-sented to his doctor’s office with a history of chronicleft-sided epiphora and a 5-day history of progressive leftperiorbital swelling that had started with a “bump” on theleft side of his nose. The patient was admitted to anoutlying hospital for left-sided orbital cellulitis and wastransferred to our institution the same day for worseningcondition.

On initial examination, the patient was hemodynami-cally stable and afebrile. Best corrected visual acuity was20/50 OD and counting fingers OS. External examina-tion revealed left-sided periorbital ecchymosis andedema with a painful nodule over the left lacrimal sac; 5mm of left-sided proptosis, marked resistance to retro-

pulsion, and limited motility were also present. Therewas no relative afferent pupillary defect. Tonometry byapplanation revealed intraocular pressures of 18 mm HgOD and 55 mm Hg OS. Anterior segment and dilatedfundus examinations were normal. CT of the orbitsrevealed left-sided preseptal and postseptal inflamma-tion, along with marked thickening of the left superiorophthalmic vein (Fig. 1). Orbital MRI with gadoliniumenhancement and fat suppression revealed a low-inten-sity signal in the left superior ophthalmic vein, consistentwith a superior ophthalmic vein thrombosis (Fig. 2).There was no cavernous sinus involvement. Blood cul-tures grew out methicillin-resistant Staphylococcus au-reus.

Treatment was initiated with intravenous vancomycinon hospital day 1, followed by a left-sided dacryocysto-rhinostomy on hospital day 3. Postoperative intravenousdexamethasone dramatically reduced the periocularswelling and improved extraocular movements. Antico-agulation was not used. Topical aqueous suppressantsand oral acetazolamide adequately controlled the ele-vated intraocular pressure. The patient was discharged on

FIG. 1. Orbital CT with left-sided orbital inflammation and di-lated superior ophthalmic vein.

FIG. 2. MRI with gadolinium contrast and fat suppressionshowing a low-intensity signal within the left superior ophthalmicvein.

388 BRIEF REPORTS


hospital day 8 with a peripherally inserted central cath-eter for home vancomycin therapy. Follow-up at 1 weekrevealed visual acuity of 20/25 OU, no afferent defect,minimal periocular swelling, near-normal resistance toretropulsion, and full motility.

DISCUSSION

Isolated septic SOVT is rare1–4 and may be a harbin-ger of cavernous sinus thrombosis.1–3 Clinical signsinclude unilateral chemosis, proptosis, limited ocularmotility, and, typically, an unremarkable fundus exami-nation.3 In this case, an SOVT was suspected after adilated superior ophthalmic vein was shown on CT.SOVT was confirmed on postcontrast MRI with fatsuppression, which revealed a low-intensity signal withinthe superior ophthalmic vein. There was no cavernoussinus involvement.

The exact pathogenesis of SOVT is unclear. Orbitalcongestion may incite an inflammatory procoagulantcascade that culminates in thrombophlebitis with throm-bus formation; alternatively, it may be due to directvascular invasion by the pathogen.2 Offending bacteriamay be aerobic or anaerobic, with S. aureus and anaer-obic streptococci being the most common organisms.3

The traditional treatment of septic SOVT consistsprimarily of aggressive antibiotic therapy.1 Theoreti-cally, corticosteroids may reduce the inflammatory pro-coagulant cascade. However, their use is supported onlyanecdotally2 or is not recommended.1 The role of anti-coagulation is unclear.1–3 Although anticoagulants haveproved successful in the treatment of septic thrombo-phlebitis in other body areas,3 there are no controlledstudies of the use of anticoagulants in treating SOVT.1

Although life-threatening hemorrhage associated withanticoagulant therapy is rare, fatal retroperitoneal hem-orrhage associated with heparin treatment of a septicSOVT has been reported.3

In this case, an underlying dacryocystitis was thoughtto be the cause of the orbital cellulitis and SOVT.Therefore, dacryocystorhinostomy was performed earlyin the hospital course. We postulate that targeted antibi-otic therapy, early dacryocystorhinostomy, and postop-erative corticosteroids without anticoagulation provedsuccessful in the treatment of an isolated septic SOVTwhile avoiding a potential septic cavernous sinus throm-bosis in this patient.

REFERENCES1. Berenholz L, Kessler A, Shlomkovitz N, et al. Superior ophthalmic

vein thrombosis: complication of ethmoidal rhinosinusitis. ArchOtolaryngol Head Neck Surg 1998;124:95–7.

2. Grassi MA, Lee AG, Kardon R, Nerad JA. A lot of clot. SurvOpthalmol 2003;48:555–61.

3. Walker JC, Sandhu A, Pietras G. Septic superior ophthalmic veinthrombosis. Clin Exp Ophthalmol 2002;30:144–6.

4. Luxenberg MN. Color Doppler imaging of superior ophthalmic veinthrombosis. Arch Ophthalmol 1991;109:582–3.

Massive Subcutaneous EmphysemaMimicking Necrotizing FasciitisAfter DacryocystorhinostomyFaris R. Ghosheh, M.D., and Sajeev S. Kathuria, M.D.

Abstract: We present a case of massive subcutaneousemphysema mimicking necrotizing fasciitis after un-complicated dacryocystorhinostomy surgery. Subcu-taneous emphysema progressing down fascial planesof the head and neck after dacryocystorhinostomyhas not been reported in the literature. Using theclinical presentation and radiographic imaging, webriefly review the underlying cause of subcutaneousemphysema after dacryocystorhinostomy.

S ubcutaneous air of the eyelids and face is rarelyseen in the postoperative setting. Orbital emphy-

sema is seen commonly, especially after trauma to theorbital floor or medial wall.1 Isolated reports of orbitalemphysema have also been noted after dental proce-dures.2 We present a case of subcutaneous emphysemaafter dacryocystorhinostomy (DCR) that involved theeyelids, face, and neck in a manner mimicking necrotiz-ing fasciitis.

CASE REPORT

A 43-year-old woman with a history of diabetes andchronic obstructive pulmonary disease (COPD) was re-ferred to the emergency room by an outside physicianwith the diagnosis of necrotizing fasciitis. She had severepain in her right jaw, difficulty chewing, and eyelidswelling. The patient had undergone uncomplicatedDCR by the senior author the day before presentation.Several hours after discharge from the hospital, shebegan to have severe emesis and coughing.

On examination, uncorrected visual acuity was 20/20OU. The patient was noted to have severe swelling andcrepitus of the right side of her face. One hour later, thesubcutaneous emphysema had advanced down to thebase of her neck. There was no erythema, tissue necrosis,or discoloration of the skin. CT of the head and neck(Figs. 1 and 2) showed significant subcutaneous emphy-sema extending from the orbit down through the face andin the neck. No focal abscess was noted.

The patient was admitted to the hospital for observa-tion. She was placed on antiemetics and told to avoidblowing her nose. She was also given two doses ofclindamycin. After observation for 1 night, the patientwas discharged. Over the next 2 weeks, the subcutaneous

Department of Ophthalmology, University of Maryland, Baltimore.Accepted for publication March 7, 2005.Address correspondence and reprint requests to Dr Sajeev S. Kathu-

ria, Department of Ophthalmology, University of Maryland, Suite 420,419 West Redwood Street, Baltimore, MD 21201.

DOI: 10.1097/01.iop.0000176267.81599.d8

389BRIEF REPORTS


emphysema resolved completely. The patient has beencompletely asymptomatic since that time.

DISCUSSION

Orbital emphysema has been described in severalsettings. Most commonly, it is encountered after orbitalblowout and medial wall fractures. Occasionally, orbitalemphysema can progress to subcutaneous emphysema.Although trauma is the most common source of subcu-taneous emphysema, there are a few postoperative casesdescribed in the literature, including after DCR.3

Necrotizing fasciitis is a progressive, rapidly spread-ing, inflammatory infection located in the deep fascia,with secondary necrosis of the subcutaneous tissues. The

presence of gas-forming bacteria results in subcutaneousair, which can be appreciated both clinically and radio-graphically. The speed of spread is directly proportionalto the thickness of the subcutaneous layer. The causativebacteria is typically group A Streptococcus but may beaerobic, anaerobic, or mixed flora. These infections canbe difficult to recognize in their early stages, but theyrapidly progress. They require early aggressive treatmentto combat the associated high morbidity and mortalityrates.

Wojno and Walter3 described a patient who had sub-cutaneous emphysema and crepitus of the eyelids afterDCR. The patient had laryngospasm after surgery andrequired manual ventilation with a face mask. In thiscase, the high positive pressure used to overcome thelaryngospasm forced air into the nasal cavity and throughthe DCR ostium, resulting in a high enough pressure tocause subcutaneous emphysema. As in our case, thispatient responded well to bed rest and avoidance of noseblowing, sneezing, or strenuous activity. No permanentsequelae were noted by the investigators.

Katz et al.4 described a series of patients who hadpostanesthesia orbital emphysema after orbital decom-pression. They noted that patients with a history ofemphysema were most likely to have bouts of coughingand Valsalva maneuvers resulting in acute orbital em-physema. However, none of the patients in this series hadlong-term visual complications.

The fascial layers of the face are collectively known asthe superficial musculoaponeurotic system (SMAS). TheSMAS divides the subcutaneous fat into two layers. Itcontains fibrous septae that extend through the fat andattach to the overlying dermis. The SMAS extends fromthe malar region superiorly to become continuous withthe galea, inferiorly to become part of the platysma, andlaterally to invest in the parotid fascia over the parotidgland.5

After DCR, patients are instructed to avoid activitiessuch as nose-blowing that create positive pressure in thenasal cavity. In our patient’s case, repeated retching dueto general anesthesia and poorly controlled emphysemaled to high positive pressure through the DCR ostiumand in the subcutaneous tissues. The SMAS then pro-vided a pathway for the air to dissect down the face andin the neck. A one-way “valve” was thus created, allow-ing increasing amounts of air to accumulate in the tis-sues. The cycle was broken by antiemetics and bed restand the “valve” was allowed to close spontaneously.Bacteria, like subcutaneous air, can also track along theSMAS. Confusion with necrotizing fasciitis by the refer-ring physician is therefore understandable in this case.

We are not aware of any other published report ofsubcutaneous emphysema of this magnitude after DCR.Although it is a rare event, ophthalmologists should beaware of the potential for subcutaneous emphysema after

FIG. 1. CT of neck (axial view). Subcutaneous emphysema inneck noted at C6 vertebra.

FIG. 2. CT of the head (coronal view). A large area of subcu-taneous air can be visualized within the orbit and extendingdown in face.

390 BRIEF REPORTS


DCR surgery. Equally important, however, is to rule outinfection by gas-producing bacteria. Careful monitoringand supportive care of the patient with massive subcuta-neous emphysema is usually all that is needed. If aninfection is suspected, appropriate treatment with antibi-otics is warranted.

REFERENCES1. Zimmer-Galler IE, Bartley GB. Orbital emphysema: case reports

and review of the literature. Mayo Clin Proc 1994;69:115–21.2. Buckley MJ, Turvey TA, Schumann SP, Grimson BS. Orbital

emphysema causing vision loss after dental extraction. J Am DentAssoc 1990;120:421–4.

3. Wojno TH, Walter K. Subcutaneous emphysema of the eyelids afterdacryocystorhinostomy. Am J Ophthalmol 1993;115:671–2.

4. Katz SE, Lubow M, Jacoby J. Suck and spit, don’t blow: orbitalemphysema after decompression surgery. Ophthalmology 1999;106:1303–5.

5. Gosain AK, Yousif NJ, Madiedo G, et al Surgical anatomy of theSMAS: a reinvestigation. Plast Reconstr Surg 1993;92:1254–65.

“Pseudo-pseudochalazion”: GiantChalazion Mimicking EyelidNeoplasmLily Koo, M.D., Mark P. Hatton, M.D., andPeter A. D. Rubin, M.D.

Abstract: A 33-year-old man presented with a solidlesion encompassing the entire left upper eyelid. Mul-tiple biopsies revealed lipogranuloma consistent withchalazion. The induration resolved after multiple tri-amcinolone injections. This is the only case report toour knowledge of a chalazion that involved the entireupper eyelid.

CASE REPORT

A 33-year-old man presented to an outside ophthal-mologist with a 3-week history of a left upper eyelidlesion that did not improve with warm compress appli-cation. He was diagnosed with a “large chalazion” of theleft upper eyelid and underwent incision and curettage.After the procedure, he used fluorometholone and eryth-romycin ointment for 5 days. He returned 2 weeks laterwith increased swelling and erythema of the left uppereyelid and was prescribed multiple courses of antibioticsincluding oral cephalexin, amoxicillin/clavulanate, andintravenous gatifloxacin. Despite 3 weeks of antibiotictherapy, the eyelid fullness did not resolve. CT revealed

no fluid collection or abscess of the eyelid. He was thenreferred to our institution.

Our initial examination revealed normal visual acu-ities and extraocular motility OU. There was completeptosis OS with no measurable levator function. Externalexamination revealed fullness of the entire left uppereyelid without overlying skin erythema or edema (Fig.1). On palpation, there was a firm mass encompassingthe entire left upper eyelid. There was no regionallymphadenopathy. There was no evidence of meibomitis,blepharitis, or dermatitis. The remainder of the anteriorsegment and dilated fundus examinations were unre-markable.

Given the concern of possible malignancy, the patientunderwent transconjunctival biopsy of the lesion. Atsurgery, a dense, solid, nondraining mass was encoun-tered. When frozen-section pathology demonstrated li-pogranuloma consistent with chalazion, an additionalexternal biopsy was performed in search of neoplastictissue (Fig. 2). These biopsies also confirmed the diagnosisof chalazion. Acid-fast bacilli and Giemsa staining were

Massachusetts Eye and Ear Infirmary, Boston, Massachusetts,U.S.A.

Accepted for publication March 11, 2005.Address correspondence and reprint requests to Dr. Peter A.D.

Rubin, Eye Plastics, Orbit, and Cosmetic Surgery, Massachusetts Eyeand Ear Infirmary, 243 Charles Street, Boston, MA 02114.

E-mail: [email protected].

DOI: 10.1097/01.iop.0000176266.14388.6b

FIG. 1. External photograph at presentation demonstratingptosis and edema of the left upper eyelid.

FIG. 2. Histopathology demonstrating lipogranuloma.

391BRIEF REPORTS


negative. The patient was started on a course of oralprednisone 1 month after biopsy. Systemic complications(anxiety) limited the dosing to 11 days. There was mildimprovement in the left upper eyelid fullness while thepatient was taking prednisone. The patient was tried on oralibuprofen without effect. Two months after presentation, 20mg triamcinolone acetonide was injected in the left uppereyelid, which resulted in moderate reduction of the fullnessof the upper eyelid. He received repeat injections of triam-cinolone acetonide (40 mg) across the length of the eyelid4 months and 14 months after the initial injection, resultingin further improvement. As the swelling resolved, a small,deep nodule was palpable in the eyelid. This was excised 7months after the last steroid injection, and pathology wasconsistent with chalazion. There was no further swellingnoted at last follow-up, 1 month after surgery (Fig. 3).

DISCUSSION

Given the atypical mass enlargement of the uppereyelid after incision and curettage of what appeared to bean ordinary chalazion, it was obvious to us to pursuehistopathologic examination of the eyelid mass to ruleout entities that masquerade as chalazia, the most impor-tant of these being neoplastic.

Sebaceous carcinoma is most frequently misdiagnosedclinically and histopathologically. It can present as aneyelid mass, recurrent chalazion, or diffuse unilateralblepharoconjunctivitis, and it is important to examineeach layer of the eyelid histologically, as was done in thiscase. One study reported 50% cases misdiagnosed.1 Al-though rare, it can be highly malignant, infiltrative, andmetastasize and has an associated mortality rate of 22%to 30%.2

Other diagnoses documented in the literature that canappear to be chalazia are Merkel cell tumor of eyelid,3

cutaneous leishmaniasis,4 tuberculosis,5 sarcoid,6 metas-tases,7 eosinophilic granuloma,8 and pilomatrixoma.9

These diagnoses are rare, and most of these causes can beruled out by histopathology.

Once the diagnosis of lipogranuloma was confirmedwith histopathology, it became an unusual problem totreat, given the size of the lesion and that it had grown insize after incision and curettage. The patient was tried onoral prednisone, and, although improved, could not tol-erate the systemic side effects. The lesion was theninjected with triamcinolone acetonide. There have beenmany successful reports with intralesional injection oftriamcinolone acetonide of chalazia. Mohan et al.10 had asuccess rate of 92.3% in 110 cases. Knowing the enor-mity of this patient’s lesion, surgical management wouldhave resulted in permanent disfigurement of the eyelid;thus, nonsurgical methods were pursued.

Although the histology was consistent with chalazion,the presentation was not. Given the absence of eyelidmarginal inflammation, blepharitis, or other signs ofinflammation in the eyelids, lack of prior chalazia, andpoor response to antibiotic therapy, the leading diagnos-tic possibilities in this case were neoplastic, but otherinfectious entities could not be excluded. The excellentresponse to steroid injections, typical of lipogranulomaand characteristic of chalazia, favor a primary inflamma-tory lesion. The immense size of this lesion remains anenigma, and we could find no similar cases in theliterature.

REFERENCES1. Khan JA, Doane JF, Grove AS Jr. Sebaceous and meibomian

carcinomas of the eyelid. Recognition, diagnosis, and manage-ment. Ophthal Plast Reconstr Surg 1991;7:61–6.

2. Yeatts RP, Waller RR. Sebaceous carcinoma of the eyelid: pitfallsin diagnosis. Ophthal Plast Reconstr Surg 1985;1:35–42.

3. Di Maria A, Carnevali L, Redaelli C, Trimarchi F. Primaryneuroendocrine carcinoma (‘Merkel cell tumor’) of the eyelid: areport of two cases. Orbit 2000;19:171–7.

4. Charif Chefchaouni M, Lamrani R, Benjelloune A, et al. Cutane-ous leishmaniasis of the lid. J Fr Ophthalmol 2002;25:522–6.

5. Aoki M, Kawana S. Bilateral chalazia of the lower eyelids asso-ciated with pulmonary tuberculosis. Acta Derm Venereol 2002;82:386–7.

6. Immonen I, Friberg K, Gronhagen-Riska C, et al. Angiotensin-converting enzyme in sarcoid and chalazion granulomas of theconjunctiva. Acta Ophthalmol (Copenh) 1986;64:519–21.

7. Mansour AM, Hidayat AA. Metastatic eyelid disease. Ophthalmol-ogy 1987;94:667–70.

8. Weissgold DJ, Wulc AE, Frayer WC, Young M. Eosinophilicgranuloma of the eyelid. Ophthal Plast Reconstr Surg 1994;10:160–2.

9. Seitz B, Holbach LM, Naumann GO. Pilomatrixoma of the eyelids:clinical differential diagnosis and follow-up: report of 17 patients.Ophthalmologie 1993;90:746–9.

10. Mohan K, Dhir SP, Munjal VP, Jain IS. The use of intralesionalsteroids in the treatment of chalazion. Ann Ophthalmol 1986;18:158–60.

FIG. 3. External photograph 3 months after steroid injectionsdemonstrating improvement of the edema and ptosis.

392 BRIEF REPORTS


Intraconal Grease-Gun Injury:A Therapeutic DilemmaFlorian Gekeler, M.D.*,Antonio A. V. Cruz, M.D., Ph.D.†,Sheila Andrade de Paula, M.D.†,Antonio Carlos dos Santos, M.D., Ph.D.‡, andFernando Chahud, M.D.§

Abstract: The case of a 31-year-old man with anaccidental grease-gun injury to his left orbital region ispresented. CT and MRI showed a well-delineated intra-conal cyst in the superior aspect of the left orbit sur-rounded by a significant inflammatory response. Thepatient was followed for 11 months. Diplopia was notreported at any time, and visual acuity, visual fields, andall other examinations remained normal. Uneventfulsurgical removal was performed at the patient’s re-quest. Histopathologic examination demonstrated a typ-ical picture of lipogranulomatous inflammation. In areview of the accessible literature, only four cases ofintraorbital grease-gun injuries were detected; all ofthem submitted to early surgical removal. We conclude,however, that an intraconal oil/grease cyst can be care-fully monitored and—in the absence of symptoms—must not necessarily be removed.

L ubricants include a wide variety of substances suchas materials designed to reduce friction between

moving surfaces or electric insulation, mold-removingagents, hydraulic fluids, coolants, and so forth. The termgrease means a lubricant in gel form containing a base oillubricant and a gelling agent. Grease-guns are habituallyused in heavy duty industry for lubrication of rotatingparts. High-pressure injuries from grease-guns have beenreported in various parts of the body including hands,genitals, and, in rare instances, in the eye and its ad-nexa.1–3 Among the wide variety of orbital foreign bod-ies, oil is the least reported. Its tolerance for extendedperiods of time is unknown. We report a case of anintraconal grease-gun injury that was followed for 11months without visual loss or other symptoms. Surgicalremoval was performed at the patient’s request.

CASE REPORT

A 31-year-old worker was referred to our hospitalafter an accidental occupational injury from a grease-gun

to his left orbital region 3 days earlier. On presentation,he complained of pain and reduced visual acuity OS.Marked edema of the upper and lower eyelids was noted.A laceration on the upper eyelid appeared to be clean andto have been adequately closed with sutures.

Biomicroscopy revealed cells in the anterior chamber.Fundus examination disclosed peripheral vitreous hem-orrhage and retinal edema in the inferior temporal quad-rant. Ocular motility was norma, and no diplopia wasreported. Visual acuity was 20/40 OS and 20/20 OD.Intraocular pressure could not be measured. All otherexaminations were normal. The working diagnosis waseyelid laceration with blunt trauma of the globe. Thepatient was treated with steroid and antibiotic eye drops.Intraocular symptoms improved quickly, with full visualacuity being reached after 6 weeks and clearing of thevitreous hemorrhage after 11 weeks.

Orbital CT 1 month after the accident showed a largecyst located in the anterior portion of the intraconal spacebetween the superior rectus muscle and the optic nerve,surrounded by an inflammatory response involving thesuperior complex. Follow-up CT and MRI (Fig. 1) after8 months showed that the cyst’s size had not changed butthe inflammatory response was reduced. Clinical exam-ination including visual acuity and visual fields, opticnerve head, and motility remained completely normalduring follow-up, with the exception of persistent palpe-bral edema.

At the patient’s request, uneventful surgical removalof the cyst’s contents and its wall was performed 11months after the accident, using an eyelid crease ap-proach. The palpebral wound was thoroughly cleaned.Histologic examinations of the biopsy specimens showedmany foci of granulomatous inflammation surroundingclear spaces in the eyelid. The cyst’s wall was formed bythick layers of collagen fibers with lymphoid aggregates(Fig. 2).

Follow-up was unremarkable, and the patient’s condi-tion is stable, with persistent but markedly reducedswelling of the left upper eyelid.

DISCUSSION

Orbital foreign bodies can cause primary mechanicalinjury to the eye, extraocular muscles, optic nerve, andbony orbit. They can also cause secondary harm due tocompression of orbital contents, mechanical restrictionof ocular motility, migration, infection, and toxicity. Thecomposition of a foreign body is a significant determin-ing factor for its tolerance. Some orbital foreign bodiessuch as most metals and steel are inert and can be welltolerated and must not necessarily be removed. Othermaterials such as wood or copper can lead to chronicinflammation, extensive scarring, and fistula formation.4

Experience with intraorbital oil is limited and in theaccessible literature, only four cases of grease-gun injuries

*University Eye Hospital, Tubingen, Germany; †Department ofOphthalmology, Otorhinolaryngology, and Head and Neck Surgery,the ‡Department of Radiology, and the §Department of Pathology,School of Medicine of Ribeirao Preto, University of Sao Paulo,Brazil.

Accepted for publication February 7, 2005.Address correspondence and reprint requests to Prof. Dr. Antonio

Augusto V. Cruz, Departamento de Oftalmologia, Otorrinolaringologiae Cirurgia de Cabeca e Pescoco, Av. Bandeirantes, 3900, 14049–900Ribeirao Preto, SP, Brazil. E-mail: [email protected]

DOI: 10.1097/01.iop.0000175017.30029.f2

393BRIEF REPORTS


to the orbit have been reported.1–3 The management ofintraconal deposition of any lubricant agent (oil or grease) isopen to debate. The composition of greases and oils ishighly variable and may contain many different classes ofadditives including polymers, alcohols, ethers, phenols, ac-ids, esters, amines, amides, sulphur, selenium, tellurium,silicon, molybdenum, phosphorus, and so forth.

Tissue reaction to the presence of any oily substance isprobably more influenced by the vehicle (the oil) than by itsadditives. In fact, the same histopathologic changes, de-scribed as a “Swiss cheese appearance,” have been reportedin response to the presence of different materials such asparaffin, petrolatum, lanolin, and distinct types of oil. Thiscondition is characterized by cystic spaces in soft tissuessurrounded by collagen and inflammatory cells, mainlylymphocytes, epithelioid cells, and multinucleated giantcells (Fig. 2). The cysts may contain clusters of sphericalbodies that represent altered red cells. In these cases, thespherules may be confused with fungi.

Different terms have been used in the literature todescribe the inflammatory changes caused by oil. Thecases reported, occurring in the bulbar conjunctiva, lac-rimal system as complications of septorhinoplasty, sinussurgeries, portal vein embolization, cortical block grafting,transconjunctival blepharoplasty, dacryocystorhinostomy,and cosmetic surgeries have been referred to as sclerosinglipogranulomatosis, lipogranuloma, paraffinoma, oilgranuloma, myospherulosis, and spherulocystosis.5

Most authors favor the surgical removal of the af-fected tissues as the only possible treatment. However, incases of intraconal oil deposition, this management isopen to debate. It is not clear if these granulomas areharmful to the orbital contents in the absence of mechan-

FIG. 1. Top: Axial (left) and coronal (right) CT scans 8 months after the accident showing an intraconal cyst in the left orbit with aresidual inflammatory response surrounding the cyst. Bottom: (left) T1-weighted MRI without contrast: low saturation of the cyst’scontents. With fat saturation and administration of gadolinium (right) the contents of the cyst show high saturation.

FIG. 2. Hematoxylin and eosin staining. A, preseptal tissues(40�) showing well-formed granuloma (arrow head) and manyclear spaces filled with grease; B, cyst wall (200�) showing alymphoid aggregate (arrowhead) composed of mature lympho-cytes on a background of dense collagen fibers.

394 BRIEF REPORTS


ical problems such as optic nerve damage or diplopia.Our case was followed for 11 months, showing noconsequences to optic nerve function. The inflammatoryresponse in the superior complex was markedly reduced,whereas the cyst size and location remained unchanged.Weighing the potential risks of an intraorbital procedureagainst damage to the orbital contents by the persistingforeign body is a therapeutic dilemma. In our case, wedecided to not operate and to carefully monitor thepatient for signs of optic nerve damage. Surgery wasperformed 11 months after the accident because thepatient was concerned and he specifically requested it. Inconclusion, we think that our case suggests that smallamounts of intraconal deposition of oily substances canbe carefully observed for extended periods of time andmay not necessarily require surgical intervention.

REFERENCES1. Dallas NL. Chronic granuloma of the orbit caused by grease-gun

injury. Br J Ophthalmol 1964;48:158–9.2. Boukes RJ, Stilma JS, de Slegte RG, Zonneveld FW. Grease-gun

injury of the orbit: computed tomography and magnetic resonanceimaging in diagnosis and treatment. Doc Ophthalmol 1987;67:273–80.

3. Goel N, Johnson R, Phillips M, Westra I. Grease gun injuries to theorbit and adnexa. Ophthal Plast Reconstr Surg 1994;10:211–5.

4. Cooper WC, Haik BG, Brazzo BG, In: Nesi FA, Lisman RD, LevineMR, eds. Smith’s Opthalmic Plastic and Reconstructive Surgery.2nd ed. St. Louis, MO: Mosby, 1998:260–8.

5. Rettinger G, Helmuth S. Lipogranulomas as complications of sep-torhinoplasty. Arch Otol Head Neck Surg 1997;123:809–14.

Granular Cell Tumor of the Orbit:Magnetic Resonance ImagingCharacteristicsMichael Ahdoot, M.D., and I. Rand Rodgers, M.D.

Abstract: A 56-year-old woman presented with choroi-dal folds and was found to have a large intraconal mass.MRI disclosed the mass to enhance with gadolinium andbe hypointense on both T1- and T2-weighted images.The tumor proved to be a granular cell tumor. Granu-lar cell tumors are rare neoplasms that may affect theorbit, lacrimal apparatus, conjunctiva, and caruncle.This is the first report describing the MRI characteris-tics of orbital granular cell tumors.

CASE REPORT

A n asymptomatic 56-year-old white woman was re-ferred by her private ophthalmologist to the oculo-

plastic service for the evaluation of choroidal striae in herleft eye. Her medical history was significant only for os-teoporosis. Her ocular history was unremarkable.

Uncorrected Snellen visual acuity was 20/20 OD and20/200, improving to 20/40 with pinhole, OS. The pupilswere equally round and reactive, with no afferent pupillarydefect. Motility examination revealed limitation of left su-praduction, and Hertel exophthalmometry was significantfor 2 mm of left globe prominence. Intraocular pressures byGoldman applanation tonometry were 18 mm Hg bilater-ally. Both optic nerves were flat with distinct margins, andleft choroidal striae were again noted.

Imaging of the orbits (Figs. 1 through 3) disclosed asharply circumscribed, ovoid 20 � 20 � 15-mm leftintraconal mass abutting the superior and lateral rectusmuscles. The lesion was hypointense on T1 weightingand homogeneously enhanced after gadolinium. On T2

weighting, the lesion remained hypointense. Sagittal sec-tions showed the mass to be inseparable from the poste-rior margin of the globe and immediately adjacent to theoptic nerve.

The patient underwent an excision of the lesion via alateral orbitotomy. During surgery, a firm, gray-coloredmass appeared well defined albeit adherent to the surround-ing orbital structures, including the superior and lateralrectus muscles. The mass was noted to encroach the opticnerve. The lesion was removed with careful dissection.

On gross inspection, the specimen appeared gray-tanin color and well encapsulated. On histologic examina-tion, the tumor cells were arranged in clusters betweenstrands of collagenous tissue and skeletal muscle cells.The tumor cells were round to oval-shaped with smallbasophilic, centrally located nuclei that lacked mitoticactivity amidst an abundant granular eosinophilic cyto-plasm (Fig. 4). The cytoplasm stained positive withperiodic acid-Schiff, Sudan black B, Indian red, andMasson trichrome. The cytoplasm also stained positivefor the �-subunit of the S-100 protein and was not altered

Department of Ophthalmology, North Shore University Hospital,New York University School of Medicine, Great Neck, New York, U.S.A.

Address correspondence and reprint requests to Dr. I. Rand Rodgers,3003 New Hyde Park Rd, Suite 409, New Hyde Park, NY 11042.

Accepted for publication January 31, 2005.All authors involved in this report have no financial interest of any

type related to the manuscript, including stock or ownership of abusiness entity connected to a product described in the paper, paidconsulting for the company or competing companies, or patent rights toa drug or piece of equipment.

DOI: 10.1097/01.iop.0000173193.56711.e2

FIG. 1. Gadolinium-enhanced T1-weighted coronal MRI scandemonstrating a hypointense left intraconal mass abutting thelateral rectus and superior rectus muscles.

395BRIEF REPORTS


by diastase predigestion. These histologic findings areconsistent with granular cell tumor.2 In addition, thepathologic margins of the tumor were clear, with noevidence of perineural spread.

Fifteen months after excision, the patient remainedasymptomatic. Her uncorrected visual acuity had im-proved to 20/30 OS. There remained a slight albeitimproved limitation of supraduction. Postoperative CTrevealed only the presence of typical postoperativechanges with an intact globe, optic nerve, and extraocu-lar muscles. No evidence of recurrence was noted.

DISCUSSION

Granular cell tumor (GCT) was originally describedby Abrikossoff1 in 1926. It has been reported throughoutthe body, with the most frequent sites of involvementbeing the tongue and subcutaneous tissues. Its occur-rence in the orbit and ocular adnexa is rare.2

The largest series of ocular GCT was reported by

Jaeger et al.3 in 1987. They presented 6 original patientsand reviewed an additional 25 previously reported cases.Ophthalmic sites have included the orbit, periorbital skinand eyelids, extraocular muscles, lacrimal sac, ciliarybody, conjunctiva, and caruncle.3

The histogenesis of GCT is not well understood. Variousproposed cell types of origin have included skeletal muscle,smooth muscle, fibroblast, histiocyte, undifferentiated mes-enchyme, neural, and Schwann cells.3 Since 1926, thetumor has been known by various names, mainly derivedfrom the presumed cell of origin.2 A Schwann cell originwas described in 1962 and has since gained popularitythrough support with various ultrastructural and immuno-histochemical studies.3 GCT cells and Schwann cells stainpositively for the �-subunit of the S-100 protein.3

Granular cell tumors of the orbit present commonlywith exophthalmos, diplopia, or decreased visual acuityfrom optic nerve involvement.3 Although usually benign,invasion of local orbital tissues has been demonstrated,with malignancy occurring in 1% to 3% of cases.3,4

Perineural spread of GCTs has been reported in othersites5 but was not seen in this study.

The presentation of orbital GCT is shared by manyorbital lesions and thus radiologic characteristics may behelpful in narrowing the differential diagnosis. Our casedemonstrates that GCTs are hypointense on T1-weightedimaging but enhance with gadolinium. On T2-weightedimaging, unlike cavernous hemangioma, hemangioperi-cytoma, schwannoma, and fibrous histiocytoma, the le-sion remains hypointense. These MRI findings are con-sistent with GCTs reported in other anatomic sites.6

Lymphomas, melanomas, and some metastatic lesionsmay also appear hypointense on T2 imaging, yet theirmorphologies may differ.

Granular cell tumors should be considered in all pa-tients presenting with slowly progressing orbital masses.

FIG. 2. On T2-weighted MRI scan, the lesion remains hypointense.

FIG. 3. Noncontrast coronal CT scan of left orbit. The mass isinseparable from the superior and lateral rectus muscles.

FIG. 4. Oval-shaped tumor cells with small basophilic nucleiarranged in clusters between strands of normal skeletal muscle(hematoxylin and eosin stain, magnification �100). Abundantgranular eosinophilic cytoplasm stained positive for periodicacid Schiff and S-100 protein immunostain.

396 BRIEF REPORTS


Our report is the first to describe the MRI characteristicsof orbital GCT and may be useful in better defining thisuncommon orbital neoplasm.

REFERENCES1. Abrikossoff A. Uber Myome, ausgehend von der quergestreiften

willkulrlichen Muskulatur. Virchows Arch 1926; (A)260: 215–33.2. Netland PA, Font RL, Jakobiec FA. Granular cell tumor. In: Albert

D, Jakobiec F, eds. Principles and Practice of Ophthalmology.Philadelphia, PA: WB Saunders Company; 1994:2055–6.

3. Jaeger MJ, Green WR, Miller NR, Harris GJ. Granular cell tumor ofthe orbit and ocular adnexae. Surv Ophthalmol 1987;31:417–23.

4. Armin A, Conelly E, Rawden G. An immunoperoxidase investiga-tion of S-100 protein in granular cell myoblastomas: Evidence forSchwann cell derivation. Am J Clin Pathol 1983;79:37.

5. Boulos R, Marsot-Dupuch K, De Saint-Maur P et al. Granular celltumor of the palate: a case report. Am J Neuroradiol 2002;23:850–4.

6. Takayama Y, Hasuo K, Takahashi N et al. Granular cell tumorpresenting as an intradural extramedullary tumor. Clin Imaging2004;28:271–3.

Advanced Periocular, Facial, andOral AmyloidosisFrancesco P. Bernardini, M.D.*,Susan Schneider, M.D.†, Carlo de Conciliis, M.D.‡,and Martin H. Devoto, M.D.§

Abstract: A 57-year-old white man presented withextensive bilateral, symmetric, confluent papules involv-ing the upper and lower eyelids, causing visual impair-ment and cosmetic deformity. Surgical debulking of thepapules was initially performed, but the lesion rapidlyrecurred and enlarged. Histopathologic examination re-vealed cutaneous amyloidosis. Six months later, exten-sive excision of the upper eyelid lesions was required torestore normal visual function.

A myloidosis is a disease characterized by the depo-sition of amyloid, a pathologic proteinaceous inter-

cellular substance in various tissues of the body; amy-loidosis can be a primary process or may occursecondary to chronic disease. The primary forms arelocalized and systemic. Primary localized amyloidosis(a.k.a. tumor-forming amyloidosis) of the ocular adnexais predominantly unilateral and tends to involve theconjunctiva and spare the eyelid skin. The involvementof the eyelid skin is suggestive of systemic amyloidosis,

although even localized amyloidosis may affect the skinin the periocular region.

Primary systemic amyloidosis, when involving theeyelid skin, usually appears as bilateral, symmetric, andconfluent papules with waxy appearance and is oftenaccompanied by purpura; the conjunctiva is usuallyspared. The association of eyelid skin lesions with mul-tiple myeloma is pathognomonic of the AL (immuno-globulin light chain) form of primary systemic amyloid-osis. In this entity, peripheral muscles, skin, nerves, andblood vessels are usually involved, and the disease canbe rapidly fatal, with a mean survival of less than 2 yearsfrom the time of diagnosis. Treatment of periocularamyloidosis remains controversial. In primary localizedamyloidosis, simple excision of the conjunctival lesionsor debulking of the suborbicularis amyloid deposits witha spooned curette have been advocated,1,2 whereas morecomplex excision followed by eyelid reconstruction hasbeen necessary for the treatment of lesions involving theeyelid margin.3,4 Multiple myeloma–associated amyloid-osis is rapidly fatal, and local treatment is therefore notrecommended unless the process causes severe cosmeticdeformity or interferes with visual function. We report acase of multiple myeloma associated systemic amyloid-osis with periocular skin lesions causing mechanicalptosis, visual field restriction, and disfiguration.

CASE REPORT

A 57-year-old man presented with multiple periocularskin lesions infiltrating the medial third of both upperand lower eyelids. The lesions were present for morethan 1 year and progressively enlarged over a 1-yearperiod. Medical history was significant for multiple my-eloma and an ischemic heart attack 10 years beforepresentation. Ocular examination revealed the visualacuity, pupillary reaction, and intraocular pressure to bewithin normal limits. Slit-lamp evaluation of the anteriorsegment and funduscopic examination of the retina were

*Department of Ophthalmology, University of Genova, Genova,Italy, and Evangelico-Valdese Hospital, Torino, Italy; †Department ofOphthalmology and Pathology and Laboratory Medicine, The NewYork Eye and Ear Infirmary, New York, New York, U.S.A.; ‡ChirurgiaOftalmoplastica, Milano, Italy; and §Consultores Oftalmologicos, Bue-nos Aires, Argentina.

Accepted for publication March 21, 2005.Address correspondence and reprint requests to Dr. Francesco P.

Bernardini, Via Zara 8, 16145 Genova, Italy. E-mail: [email protected]..

DOI: 10.1097/01.iop.0000176261.81284.cbFIG. 1. Preoperative appearance of the amyloid deposits in theperiocular region (A) and in the inner surface of the cheek (B).

397BRIEF REPORTS


unremarkable. External examination revealed the pres-ence of eyelid skin nodules of various sizes infiltratingboth upper and lower eyelids (Fig. 1A). The diagnosis ofamyloidosis was suspected, and surgical debulking of thelesions was performed. Histopathologic examination ofthe resected specimen revealed the presence of slightlyeosinophilic, amorphous material, consistent with amy-loidosis was confirmed by positive Congo red staining(Fig. 2). After moderate temporary subjective improve-ment, the lesions recurred and progressed over a periodof 8 to 10 months, causing visual field impairment,increased eyelid weight, and difficulty keeping the eyesopen throughout the day. The patient noted similar le-sions on the skin of the neck and around the nostrils.Examination revealed additional lesions on the innersurface of both cheeks filling the lateral aspect of themouth and interfering with dental occlusion (Fig. 1B).After the risks, benefits, and alternatives of surgery werediscussed with the patient, wide excision of the perioc-ular lesions was performed, proceeding with one eyelidat a time and starting with the left upper eyelid, which

was considered the worst. After removal of the affectedpart of the left upper eyelid skin, the residual defect wasreconstructed with two large retroauricular skin graftsharvested from both sides. The grafts were sutured to theresidual normal eyelid skin and joined at the upper eyelidcrease, which was reformed with deep fixation to thelevator aponeurosis. A cotton bolster was sutured inplace and removed after 1 week. Follow-up examination1 year after surgery revealed a viable graft, normal eyelidheight with no lagophthalmos, and improved cosmesis ofthe left upper eyelid. The patient is now receiving che-motherapy to treat his multiple myeloma, and furthersurgery has been postponed.

DISCUSSION

AL amyloidosis is rapidly progressive and fatal, witha mean survival rate of less than 2 years from diagnosis.Evaluation of the periocular skin lesions can help inestablishing the diagnosis. Because the prognosis is notaffected by excision, surgery is necessary only if there issevere cosmetic deformity or visual impairment. In thiscase, we planned a multistep reconstruction to improvecosmesis and visual function and to prevent pupillaryaxis occlusion by the amyloid deposition in the uppereyelid skin bilaterally.

Periocular skin lesions and macroglossia are welldocumented features of primary systemic amyloidosis,5,6

but they usually do not require surgery. To our knowl-edge, this is the first reported case of amyloidosis of theeyelid skin that required extensive surgery with skingrafting to treat visual field impairment and preventcomplete pupillary axis obstruction. The rapid progres-sion of the skin and oral lesions is also a unique charac-teristic that has not been previously reported with amy-loidosis.

REFERENCES1. Patrinely JR, Koch DD. Surgical management of advanced ocular

adnexal amyloidosis. Arch Ophthalmol 1992;110:882–5.2. Stack RR, Vote BJ, Evans JL, Elder MJ. Bilateral ptosis caused by

localized superficial eyelid amyloidosis. Ophthal Plast ReconstrSurg 2003;19:239–40.

3. Pelton RW, Desmond BP, Mamalis N, et al. Nodular cutaneousamyloid tumors of the eyelids in the absence of systemic amyloid-osis. Ophthalmic Surg Lasers 2001;32:422–4.

4. Fett DR, Putterman AM. Primary localized amyloidosis presentingas an eyelid margin tumor. Arch Ophthalmol 1986;104:584–5.

5. Laouissi N, Rais L, Zemiati S, et al. Isolated and primary palpebro-conjunctival amyloses apropos of 2 cases. J Fr Ophthalmol 1998;21:152–5.

6. Brownstein MH, Elliott R, Helwig EB. Ophthalmologic aspects ofamyloidosis. Am J Ophthalmol 1970;69:423–30.

FIG. 2. Histopathologic examination reveals the presence ofslightly eosinophilic, amorphous material, consistent with amy-loid deposition (A) and confirmed by positive Congo red stainingwith apple green birefringence (B).

398 BRIEF REPORTS


Chronic Inflammation FromPolycarbonate Motility Peg InhibitsOsteogenesis in a HumanHydroxyapatite Orbital ImplantManoj M. Thakker, M.D.*, Jing Zhang, M.D.†, andBryan S. Sires, M.D., Ph.D.*

Abstract: The histologic findings of a pegged hy-droxyapatite orbital implant removed due to chronicinflammation and pain are described. A 44-year-oldwoman underwent explantation of a hydroxyapatitesphere and polycarbonate motility peg due to chronicredness, swelling, discharge, and pain. Histology re-vealed complete fibrovascularization of the implant,with approximately 90% ossification. No bone mar-row was identified. Histologic sections revealed fi-brous connective tissue at the periphery of a scleroticbony mass with a granulomatous inflammatory infil-trate at the motility peg aperture. There were nobacterial, mycobacterial, or fungal organisms identi-fied histologically or by culture. Consistent with pre-vious reports, hydroxyapatite orbital implants areamenable to fibrovascular ingrowth and bony trans-formation. The presence of a granulomatous inflam-matory reaction around the polycarbonate motilitypeg in this case may have prevented complete osseoustransformation of the hydroxyapatite implant.

O ver the past decade, porous hydroxyapatite (HA)spheres have been increasingly used in patients

requiring orbital implants. This material is amenable tofibrovascular ingrowth, resulting in biointegration withthe surrounding host tissues.1 Theoretical advantagesthat favor the use of these biointegrated implants includea decreased risk of migration within the orbit and anincreased resistance to infection as a result of immuno-logic surveillance within the implant.2

It has also been demonstrated that HA is subject toosteogenesis and conversion in bone with hematopoieticelements.3 This probably occurs by the process of os-teoinduction, whereby osteoblast progenitor cells colo-nize the implant after vascular ingrowth has taken place.The conversion of HA into the host’s bone tissue mayalso reduce the risks of foreign body reaction and infec-tion and offer increased strength necessary to stabilizemotility coupling pegs.

In this report, we show the histologic appearance of a

human HA orbital implant removed due to chronic or-bital inflammation and pain. We demonstrate a chronic,granulomatous inflammatory reaction around the motil-ity peg site with incomplete ossification of the HA.

CASE REPORT

A 44-year-old woman with left-sided anophthalmiapresented to the University of Washington in 2003 witha 10-year history of pain, discharge, and prosthesis in-tolerance. During childhood, the patient underwent enu-cleation of the left eye for an “intraocular tumor.” Priormedical records and pathology reports were not avail-able. An HA sphere wrapped in sclera was implanted atanother institution (patient age, 34 years), and a sleevedpolycarbonate motility peg was placed 6 months later.Her symptoms began 2 months after peg placement.Before presentation, there were no cultures performed onthe implant. During this time, the patient was empiricallytreated with multiple courses of systemic antibiotics andtopical antibiotic/steroids without improvement.

Examination revealed moderate conjunctival injection,mild mucus discharge, and moderate tenderness to pal-pation. There was a stable, well-centered polycarbonatemotility peg. There was no purulent discharge, implantexposure, papillary reaction, eyelid swelling, or ery-thema.

Based on the patient’s symptoms and possibility ofindolent infection, implant removal with dermis-fat re-construction was performed. Intraoperative cultures ofthe conjunctiva and implant were obtained, and theimplant was submitted for pathologic analysis.

The HA sphere was decalcified and sectioned sagit-tally. Staining with hematoxylin and eosin, Gomorimethamine silver, periodic acid-Schiff, Brown-Brenn,and Ziehl-Nielsen was performed. Bone matrix waspresent throughout the entire implant except adjacent tothe motility peg aperture (Fig. 1, A and B). A fibrouscapsule surrounded the periphery of the implant withoccasional lymphoid germinal centers adjacent to HAmaterial (Fig. 1, C and D). Surrounding the motility pegaperture was a mixed acute and chronic inflammatoryinfiltrate with histiocytes and multinucleated cells thatmay represent giant cells or osteoclasts (Fig. 2, A and B).In the central area of inflammation, there was signifi-cantly less bone matrix and more unossified HA com-pared with the periphery of the implant (Fig. 2A). Thecollagenous fibers of the bone matrix appeared to be lessorganized within areas of inflammation (Fig. 2, C andD). Intraoperative cultures and special stains were neg-ative for bacteria and fungi.

DISCUSSION

Evidence suggests that the mechanism of bone forma-tion in porous HA implants is a gradual, multistepprocess. After implantation, serous fluid fills the porous

*Department of Ophthalmology and Department of Otolaryngology–Head and Neck Surgery, University of Washington, Seattle; and the†Department of Pathology, Harborview Medical Center, Seattle, Wash-ington, U.S.A.

Accepted for publication March 21, 2005.Address correspondence and reprint requests to Dr. Bryan S. Sires,

University of Washington, Ophthalmology, Box 356485, Universityof Washington, Seattle, WA 98195, U.S.A. E-mail: [email protected]

DOI: 10.1097/01.iop.0000179375.66916.59

399BRIEF REPORTS


cavities of the HA.4 Fibrovascular ingrowth begins inabout 2 weeks and is complete by 3 to 6 months; theprecise time period dependent on the nature of the wrap

FIG. 1. A, Explanted hydroxyapatite sphere showing motilitypeg aperture (p). Bone matrix is present except in the zoneadjacent to the peg aperture (black arrows) (hematoxylin andeosin stain, magnification 5�). B, Higher magnification viewdemonstrates replacement of hydroxyapatite with bony matrix(hematoxylin and eosin stain, magnification 20�). C, Lymphoidfollicle (white arrow) is present at the junction of the scleral wrapand implant (delineated by asterisks). Unossified hydroxyapatiteis present adjacent to the follicle (black arrow) (hematoxylin andeosin stain, magnification 10�). D, Germinal center (g) andmantle zone (m) of the lymphoid aggregation are discernable athigher magnification (hematoxylin and eosin stain, magnification40�).

FIG. 2. A, Mixed inflammatory infiltrate (asterisks) was notedadjacent to the motility peg aperture (hematoxylin and eosinstain, magnification 10�). Ratio of unossified hydroxyapatite(arrow) to bone matrix was greatest in this region. B, Highermagnification of the inflammatory infiltrate demonstrates thepresence of multinucleated cells, indicative of giant cells orosteoclasts (arrows) (hematoxylin and eosin stain, magnification40�). C, Appearance of bony matrix (m) adjacent to the zone ofgranulomatous inflammation (asterisk) near the motility peg ap-erture (hematoxylin and eosin stain, magnification 25�). Colla-gen lamellae of the bony matrix (m) in this region appear to bemore disorganized (characteristic of immature bone) when com-pared with uninflamed areas of the implant (D) (hematoxylin andeosin stain, magnification 25�).

400 BRIEF REPORTS


material and whether holes are drilled before implanta-tion.5 Circulating osteoblast progenitor cells may subse-quently gain access to and colonize the HA structure ofthe implant through this blood supply. By signalingmechanisms related to calcium phosphate and other sur-face proteins on the implant, these progenitor cells maybe stimulated to differentiate in osteoblasts that initiatebone formation.6

Animal studies show that bone formation in HA im-plants can occur as early as 6 weeks.7 In humans, thereare reported cases of HA spheres demonstrating incom-plete bone formation as early as 5 and 7 months afterimplantation.5 Another report of an explanted human HAsphere without a motility peg demonstrated that ossifi-cation of the implant is virtually complete (with bonemarrow) after 5.5 years.3

Of importance is the granulomatous inflammatory re-sponse around the motility peg site. The absence of infec-tious organisms, presence of histiocytes, and chronic natureof the inflammation suggest that this may be due to aforeign body reaction against the polycarbonate peg. In-flammatory cytokines, such as interleukin-1 and tumornecrosis factor, have been shown in animal models toinhibit osteoblasts and promote osteoclast differentiationand activity.8 This may in part explain the paucity, disor-ganization, and lack of ossification of the bone matrixsurrounding the motility peg site. Chronic exposure toinflammatory cytokines may also have similarly inhibiteddifferentiation of hematopoietic progenitor cells, resultingin the absence of bone marrow in this implant. If this typeof inflammatory/foreign body reaction is common, ourfindings may in part explain the high rate of complications(i.e., discharge, peg instability, infection) associated withthese pegs.9

REFERENCES1. Rubin PAD, Popham JK, Bilyk JR, Shore JW. Comparison of

fibrovascular ingrowth into hydroxyapatite and porous polyethyleneorbital implants. Ophthal Plast Reconstr Surg 1994;10:96–103.

2. Merritt K, Shafer JW, Brown SA. Implant site infection rates withporous and dense materials. J Biomed Mater Res 1979;13:101–8.

3. Sires BS, Benda PM. Osteogenesis in a human hydroxyapatiteorbital implant 5.5 years after implantation. Am J Ophthalmol2000;130:368–9.

4. Sires BS, Holds JB, Archer CR, et al. Histological and radiologicalanalyses of hydroxyapatite orbital implants in rabbits. Ophthal PlastReconstr Surg 1995;11:273–7.

5. Ferrone PJ, Dutton JJ. Rate of vascularization of coralline hydroxy-apatite ocular implants. Ophthalmology 1992;99:376–9.

6. Beck GR. Inorganic phospate as a signaling molecule in osteoblastdifferentiation. J Cell Biochem 2003;90:234–43.

7. Sires BS, Geggel HS, Heffernan JT, et al. Use of demineralizedbone as an osteoconductive orbital enucleation implant in the rabbit.Ophthal Plast Reconstr Surg 1993;9:112–9.

8. Walsh NC, Gravallesse EM. Bone loss in inflammatory arthritis:mechanisms and treatment strategies. Curr Opin Rheumatol 2004;16:419–27.

9. Jordan DR, Chan S, Mawn L, et al. Complications associated withpegging hydroxyapatite orbital implants. Ophthalmology 1999;106:505–12.

Delayed Inflammatory Reaction toHyaluronic Acid (Restylane�)David R. Jordan, M.D.

Abstract: A 45-year-old man received an injection ofRestylane (cross-linked hyaluronic acid) to a mid-forehead furrow line in an attempt to reduce itsprominence. The injection was uncomplicated andsuccessful in reducing the line. Five months after theinjection, he returned with an inflammatory reactionin the area of injection that resolved without treat-ment over a period of 3 weeks.

A variety of soft tissue fillers have been used todiminish facial folds and wrinkles, including sili-

cone, autologous fat, bovine and human collagen, andothers.1 A new filler popularized in recent years isinjectable hyaluronic acid (HA). Hyaluronic acid is abasic building block of the dermis that binds water andcreates volume. It exhibits no species or tissue specific-ity, and the chemical structure of this polysaccharide isuniform throughout nature. HA may be biotechnologi-cally manufactured with bacterial fermentation tech-niques (Restylane; Q-Med Aesthetics, Stockholm, Swe-den) or extracted from rooster combs (Hylaform;Biomatrix Inc., Ridgefield, NJ, U.S.A.). Manufacturersand distributors of these products suggest that there is noneed for skin testing because there is little or no potentialfor immunologic reactions in humans.1–3 However, thisdoes not mean these products are risk free, as there arepotential problems that may occur.2–9 This article reportsa delayed inflammatory reaction occurring 5 monthsafter an injection of Restylane.

CASE REPORT

A 40-year-old, previously well white man had a prom-inent horizontal forehead furrow line and the vermillionborder of his upper lip injected with Restylane from thesame syringe (June 1999). Both areas looked fine 1 weekafter injection. Five months later, he returned with anelevated, lumpy red line in the forehead that developedover a 24-hour period (Figure). The lip, however, ap-peared uninvolved. There was no pain or tenderness inthe involved area. The inflammatory reaction graduallydisappeared over the next 3 weeks.

DISCUSSION

Several authors have questioned the purported lack ofimmunogenicity to HA products, based on inflammatory

University of Ottawa Eye Institute, Ottawa, Canada.Accepted for publication January 19, 2005.Address correspondence and reprint requests to Dr. David R. Jordan,

104, 340 McLeod St, Ottawa, Ontario, Canada. E-mail: [email protected].

DOI: 10.1097/01.iop.0000173194.18050.b8

401BRIEF REPORTS


reactions3–5 and foreign body granuloma development6,7

that occurred after HA injection.Lupton and Alster3 reported a 54-year-old woman who

had development of multiple red nodules within thetreatment area 2 weeks after a third session of HAinjection (June 1999; HA type not disclosed). A potentialallergy to an impurity of the bacterial fermentation of theHA gel was the suspected cause. Any impurities presentduring the manufacturing process could account for asensitization reaction.3

Lowe et al.4 described delayed inflammatory reactionsoccurring 6 to 8 weeks after HA injection (Hylaform andRestylane) in 6 patients (1996 to 2000). An allergic reactionto the HA products was implied, but the precise nature ofthe reactivity was not determined. Furthermore, Klein,5

through personal communication with Lowe, reports Loweno longer supports the conclusion of this article as valid.

Micheels6 reported 8 patients with redness, pruritus,and painful swelling in areas of HA injection that lastedup to 5 months. He suggested these reactions may besecondary to antibodies against the HA products (IgG,IgE), but the evidence to support this was limited.5 Otherarticles7,8 reporting foreign body granuloma formationand suggesting allergic reactions to the HA productsimilarly provide no supportive evidence that the possi-ble mechanism of the reaction is allergic in nature.5

Friedman et al.9 performed a retrospective review ofthe worldwide data gathered by the manufacturer forRestylane on reported adverse events in 1999 and 2000.In 1999 there was approximately 1 adverse event re-ported for every 650 (0.15%) patients treated. The mostcommon reaction to injectable HA was a localized hy-persensitivity reaction that consisted of swelling, ery-thema, and induration at the implant site (medium dura-

tion, 15 days). There were also rare reports of localizedgranulomatous reactions, bacterial infection, and acne-iform and cystic lesions. In 2000, the total number ofadverse events was 1 for every 1,800 (0.06%) patientstreated. This decrease was likely due to a reformulationof the HA. In mid-1999, an HA raw material wasintroduced with trace amounts of protein 6 times lowerthan the raw material previously used.9 Hypersensitivityreactions to HA products were probably secondary toimpurities of bacterial fermentation.9

In the present case, the lip appeared to have noinflammatory reaction, which may be because only asmall amount of Restylane was injected along the ver-million border and a delayed inflammatory reaction wasnot clinically obvious, or because the inflammatory re-action was camouflaged by the natural redness and in-distinct outline of the patient’s vermillion border.

In summary, it is clear there may be rare inflammatoryreactions to HA products occurring 2 to 8, or even 20 weeksafter injection, as in the present case.3,4,6–9 Whether theseare a result of allergy to the HA or a reaction to theimpurities in the product has been debated.3–9 An antibodyresponse to HA itself, however, has never been docu-mented, and if there is any immunologic response to theseproducts, it is most likely a response to the proteincontaminants rather than the HA.5 The reformulation ofRestylane in 1999 resulted in a 6-fold reduction ofprotein load in the final product.5,9 This reduction hasresulted in a concomitant reduction in the reported oc-currences of suspected hypersensitivity reactions.5,9

REFERENCES1. Jordan DR. Soft tissue fillers for wrinkles, folds and volume

augmentation. Can J Ophthalmol 2003;38:285–2.2. Goa KL, Benfield P. Hyaluronic acid: A review of its pharmacology

and use as a surgical aid in ophthalmology, and its therapeuticpotential in joint disease and wound healing. Drugs 1994;47:536–66.

3. Lupton JR, Alster TS. Cutaneous hypersensitivity reaction to inject-able hyaluronic gel. Dermatol Surg 2000;26:135–7.

4. Lowe NJ, Maxwell CA, Lowe P et al . Hyaluronic acid skin fillers:adverse reactions to skin testing. J Am Acad Dermatol 2001;45:930–3.

5. Klein AW. Granulomatous foreign body reaction against hyaluronicacid (letter). Dermatol Surg 2004;30:1070–1.

6. Micheels P. Human anti-hyaluronic acid antibodies: Is it possible?Dermatol Surg 2001;27:185–91.

7. Raulin C, Greve B, Hartschuh W, Soedging K. Exudative granulo-matous reaction to hyaluronic acid (Hylaform). Contact Dermatitis2000; 43: 178–9.

8. Fernandez -Acenero MJ, Zamorra E, Borbujo J. Granulomatousforeign body reaction against hyaluronic acid: report of a case afterlip augmentation. Dermatol Surg 2003;29:1225–6.

9. Friedman PM, Mafong EN, Kauvar AN, Geronemus RG. Safetydata of injectable non animal stabilized hyaluronic acid gel for softtissue augmentation. Dermatol Surg 2002;28:491–4.

An elevated, red, lumpy line is seen in the area of a previousRestylane injection (arrows).

402 BRIEF REPORTS


Eyelid Involvement in AcanthosisNigricans: The Importance ofSystemic ScreeningDidem Serin, M.D.*, Ali Haydar Parlak, M.D.†,Gursoy Alagoz, M.D.*, Cetin Boran, M.D.†, andNadir Goksugur, M.D.†

Abstract: A 53-year-old man presented with thick-ening and hyperpigmentation of the eyelids. After thediagnosis of acanthosis nigricans was made, furtherinvestigation of a possible underlying disorder, in-cluding biochemical and instrumental examinations,indicated the presence of glucose intolerance, whichhad been diagnosed 3 years after the clinical appear-ance of acanthosis nigricans. Eyelid involvement inacanthosis nigricans is rare. Ophthalmologists shouldbe aware of the possibility that acanthosis nigricanscan exist in the periocular area and perform a sys-temic assessment of the patient for the presence of anunderlying condition.

A canthosis nigricans (AN) is characterized by skinhyperpigmentation and thickening, most commonly

involving the intertriginous areas such as the axillae,neck, groin, and external genitalia. Eyelid involvement isatypical for AN. Although AN itself is an innocuousdisorder, it is commonly a manifestation of an underlyingsystemic disease and often precedes its clinical symp-toms. AN may occur secondary to obesity, glucose in-tolerance, insulin resistance, pituitary hypogonadism,Addison disease, hyperthyroidism, hypothyroidism, anduse of nicotinic acid, corticosteroids, and insulin, or, itmay develop as a paraneoplastic syndrome accompany-ing various malignancies, particularly gastrointestinaladenocarcinomas. AN may appear before, concomitantwith, or after the clinical manifestations of the underly-ing pathology.1,2 We aim to emphasize the significanceof further investigation after the diagnosis of AN is madeby presenting a rare case with AN that is localized in andlimited to the periocular area and forehead. To ourknowledge, this is the first report of a case with such aconfined and rare localization, which revealed glucoseintolerance.

CASE REPORT

Data accumulation was approved by the ethics com-mittee of our university, and the patient gave consent. A53-year-old white man had a 3-year history of bilateralprogressive thickening and hyperpigmentation of theskin of his eyelids and forehead accompanied by papil-lomatous growths on the eyelids and a 0.5-cm cysticlesion on the left lateral canthus, clinically consistentwith AN and hydrocystoma (Fig. 1). A full examinationof the entire skin surface was performed by the derma-tologists and intertriginous areas such as axillae, neck,and inguinal regions were free from AN. The cutaneousalterations were confined to the periocular region andforehead and did not cause any functional symptomssuch as epiphora from occlusion of the canaliculi withpapillomatous lesions, but the patient complained of thehydrocystoma for cosmetic reasons. A thorough ophthal-mologic examination including pupil reactions, best-corrected visual acuity, intraocular pressure, biomicros-copy, funduscopy, ocular motility, and lacrimal drainagerevealed no other pathologic findings except presbyopia.The cystic lesion was excised, and several biopsy spec-imens adjacent to the lesion were taken. The clinicaldiagnosis of AN and hydrocystoma was confirmed bythe histopathologic study, which demonstrated hyperker-atosis, mild acanthosis, and papillomatosis in the epider-mis and a cystic space lined by a double layer of cuboidalcells (Fig. 2). The patient underwent further investigationof a possible occult disorder. The patient’s only systemiccomplaint was an abdominal ache. He had not taken anymedication for the past year and had not noted anyweight loss. Routine analyses, including full blood count,urine analysis, and thyroid function tests, were withinnormal ranges. Fecal analysis did not reveal occultblood. Biochemical tests performed twice at a 1-monthinterval revealed 125 mg/dl and 115 mg/dl fasting

*Department of Ophthalmology, Department of Dermatology, and†Department of Pathology, Izzet Baysal Medical Faculty, Bolu, Tur-key.

Accepted for publication March 21, 2005.Presented at the 38th Turkish National Ophthalmology Congress,

Antalya, Turkey, 2004.Address correspondence and reprint requests to Dr. Didem Serin,

Abant Izzet Baysal University, Izzet Baysal Medical Faculty, Depart-ment of Ophthalmology, 14280, Golkoy, Bolu, Turkey. E-mail:[email protected]

DOI: 10.1097/01.iop.0000176262.58350.38

FIG. 1. Photograph demonstrating hyperpigmentation andthickening of the patient’s left periocular area and hydrocystomaon the left lateral canthus.

403BRIEF REPORTS


plasma glucose at the first and second tests, respectively(normal range, 70 to 109 mg/dl). The oral glucose toler-ance test disclosed a plasma glucose concentration of163 mg/dl at the second hour. These impaired values ofplasma glucose led to the diagnosis of glucose intoler-ance according to the criteria of the American DiabetesAssociation. Tumor markers including prostate-specificantigen, free prostate-specific antigen, �-fetoprotein, car-cinoembryonic antigen, and CA 19,9 were also withinnormal limits. Ultrasonography of the abdomen dis-closed the presence of gallstones in his gallbladder.Chest radiography, abdominal CT, gastroduodenoscopy,and colonoscopy were otherwise unremarkable.

DISCUSSION

Atypical localizations in AN including hands, feet,lips, buccal mucosa, soft palate, sulcus mentolabialis,eyelids, and submammary region are rarely reported. Wereport this case to highlight two significant aspects ofAN. The first is the rare periocular involvement; thesecond is the importance of further investigation after thediagnosis of AN has been made.

Bottoni et al.3 diagnosed AN involving the neck,axillae, mouth, and eyelid conjunctivae and discovered anon–small cell carcinoma of the lung after a series ofinstrumental and biochemical examinations. Cutaneousalterations had appeared 1 year before the diagnosis ofthe malignancy and directed the examination that dis-

closed his lung cancer. In a report by Groos et al.,4 a casewith AN involving the eyes and many other regions ofthe body was described. AN had developed 1 year beforethe diagnosis of gastric adenocarcinoma, and the patienthad persistent epiphora from occlusion of all four punctawith papillomatous lesions accompanying AN. Othercase reports described AN as a paraneoplastic syndromethat led to diagnoses of colon and lung malignancies.5,6

In our case, all the investigations performed to demon-strate any possible associated disorder indicated the pres-ence of glucose intolerance diagnosed 3 years after theclinical appearance of AN.

In a study by Kapasi et al.,7 6 patients who presentedto the dermatology clinic with AN were evaluated to ruleout endocrine diseases, and glucose intolerance wasfound to be the associated disorder. Another study con-cluded that AN could be a reliable cutaneous marker ofinsulin resistance in obese children. Children with ANshowed significantly more glucose intolerance comparedwith those without AN.8 However, in these studies, noneof the subjects presented with periocular involvement.

AN can often precede the clinical signs of the under-lying condition. Therefore, awareness of the possibilitythat AN can exist in the periocular area may help oph-thalmologists suspect an occult endocrine or malignantdisorder. It is advisable that thorough systemic assess-ment of patients presenting with AN be performed afterits recognition.

REFERENCES1. Schwartz RA. Acanthosis nigricans. J Am Acad Dermatol 1994;31:

1–14.2. Griffiths WAD, Judge MR,LeighIM . Disorders of keratinization.

In: ChampionRH,BurtonJLBurnsDA BreatnachSM Textbook ofDermatology. 6th ed. Oxford: Blackwell Science; 1998:1483–588.

3. Bottoni U, Dianzani C, Pranteda G, et al Florid cutaneous andmucosal papillomatosis with acanthosis nigricans revealing a pri-mary lung cancer. J Eur Acad Dermatol Venereol 2000;14:205–8.

4. Groos EB, Mannis MJ, Brumley TB, Huntley AC. Eyelid involve-ment in acanthosis nigricans. Am J Ophthalmol 1993;115:42–5.

5. Wedge CCI, Rootman DS, Hunter W, Sibbald G. Malignant acan-thosis nigricans. Ophthalmology 1993;100:1590–2.

6. Pinto GL, Meyer DR. Ophthalmic manifestations of acanthosisnigricans. Ophthal Plast Reconstr Surg 1994;10:49–50.

7. Kapasi A, Varthakavi PK, Khopkar U, et al Endocrine profiles in sixpatients with acanthosis nigricans. J Assoc Physicians India 1994;42:529–30.

8. Yamazaki H, Ito S, Yoshida H. Acanthosis nigricans is a reliablecutaneous marker of insulin resistance in obese Japanese children.Pediatr Int 2003;45:701–5.

FIG. 2. Histopathology reveals papillomatosis, mild acantho-sis, hyperkeratosis, and irregular basal hyperpigmentation (he-matoxylin and eosin stain, magnification �100).

404 BRIEF REPORTS


Documents

Primary Squamous Cell Carcinoma Arising From an Epithelium ... · Squamous cell carcinoma is a rare neoplasm of the lacrimal gland and may be either primary or secondary in origin